The Golgi complex is an essential organelle involved in the post-translational modifications of proteins and in the targeting of membrane and secreted proteins to their destination in the cell. It is easily identified in electron micrographs as a stack of flattened cisternae, each representing a different functional compartment through which secreted and trans-membrane proteins transit sequentially. Yet, there is plasticity in its organization. The muscle Golgi complex is constituted of many small stacks, whereas most mammalian cells have a single Golgi complex. Its positioning near the nucleus is due to microtubules, and different microtubule organizations seem responsible for different Golgi complex organizations. We have published the only comprehensive model, so far, for the reorganization of the Golgi complex during muscle differentiation (Lu et al., 2001). During muscle differentiation, both microtubules and Golgi complex are coordinately redistributed as muscle cells elongate and fuse to form large multinucleated cells. We demonstrated that reorganization of the Golgi complex involves its fragmentation in small elements. We proposed that this reorganization depends on the constant recycling of Golgi complex proteins through the endoplasmic reticulum (ER), on the reorganization of microtubule organizing centers (MTOC, or centrosome), and on redistribution of the ER exit sites. Our current goal is to refine this model and, particularly, to find a hierarchy in the several changes that appear to be taking place simultaneously. [unreadable] We have started by asking whether reorganization of the Golgi complex can take place in the absence of a normal microtubule network. Working with the mouse muscle cell line C2, we have established concentrations of different drugs that chronically affect microtubules while allowing muscle differentiation (expression of the transcription factor myogenin). We have used the drugs nocodazole, which prevents microtubule nucleation and elongation, taxol, which instead forces massive stabilization and bundling of microtubules, and an organometallic specific inhibitor of glycogen synthase kinase 3, expected to affect microtubule orientation and stabilization.[unreadable] As we had hypothesized, differentiation in presence of these drugs affects different aspects of the cellular reorganization distinctly. Reorganization of the MTOC is the least affected, with more than 75% normal reorganization in all conditions. In contrast, Golgi complex reorganization fails or is incomplete in a large fraction of myogenin-positive cells and is never observed unless MTOC reorganization is normal. A small percentage of cells seem to have a normal reorganization of the Golgi complex without ER exit sites redistribution but in >90% of the cells the two reorganize coordinately. We therefore propose that the primary event during differentiation is the reorganization of the MTOC which then leads to the coordinate reorganization of Golgi complex and ER exit sites. Kristien Zaal is presently writing up this work.[unreadable] Although failure to reorganize completely may have physiological consequences if carried over to a mature muscle fiber, this is not necessarily the case in cultured cells. The Golgi complex may be plastic enough to function in several geometries. To assess its functionality, we have used a well-known assay, in which trafficking of a fluorescently tagged viral protein (VSV-G-GFP) is followed from the endoplasmic reticulum to the plasma membrane, through the Golgi complex. We have demonstrated that trafficking of VSV-G appears normal in cells treated with a glycogen synthase kinase inhibitor, despite the abnormal distribution of the Golgi complex elements. [unreadable] Information about the MTOC organization in muscle cells is sought through experiments carried out by Tan Zhang. He has been investigating the role of the microtubule-associated protein EB1 in muscle differentiation. In other mammalian cells, EB1 is necessary for microtubule stabilization. We hypothesized that EB1 may play a similar role during muscle development. Indeed, we found that dominant-negative constructs of EB1 affect myoblast elongation. Knocking down EB1 permanently prevents microtubule stabilization and, unexpectedly, prevents differentiation of the muscle cells (this work is described in a recently submitted manuscript). In addition to its role at the plus end tips of microtubules, EB1 is also present in the centrosome. We confirmed that EB1 knock-down cells have defects in microtubule nucleation and also found that pericentrin, one of the best known proteins of the MTOC, is partially displaced to the Golgi complex in EB1 knock-down cells. We are now ready to use cDNA constructs which affect EB1 selectively at the centrosome in order to determine which of its localizations is related to its role in cell differentiation. These results point to a link between the microtubule cytoskeleton organization and the capacity for muscle cell differentiation.[unreadable] Microtubules play an important role in the reorganization of the Golgi complex during myogenesis but they may not be the only cytoskeletal system involved in the distribution of the Golgi complex in muscle fibers and others as well as our own results suggested intermediate filament involvement. Since desmin is the main intermediate filament protein of skeletal muscle, we have used desmin-null mice. We have observed that in slow muscle fibers from desmin-null mice, Golgi complex elements and ER exit sites are abnormal and heterogeneous in size compared to those in control mice. In the past year, John Sheridan has prepared satellite cell cultures from single muscle fibers of specific muscles from control and desmin-null mice. The Golgi complex reorganization/ fragmentation in desmin-null myotubes is indeed incomplete, suggesting that desmin does plays a role in the reorganization of the Golgi complex at the early stages of differentiation.[unreadable] The best tool to obtain the information we require is immunofluorescence of single fibers. The manual teasing of the fibers is labor-intensive, yet is necessary because antibodies do not penetrate bundles of fibers and sections provide a view that is too limited. There are cases, however, where handling single fibers is close to impossible. We have been collaborating with Drs. Plotz and Raben in NIAMS who study Pompe Disease, a lysosomal storage disorder in which cardiac and skeletal muscles are the source of the pathology. Our work has emphasized the presence of large areas of autophagic debris in muscles of the mouse model and of Pompe Disease human subjects. If such buildup is linked to the pathology, it is very important to be able to assess its prevalence in biopsies, particularly in those of infants who suffer the most devastating form of the disease. Single infant fibers, however, are thin and frail, and extremely difficult to handle. To address this problem, we have used a new imaging modality which provides structural information from unstained tissue at the molecular level, the 2-photon microscopy technique called Second Harmonic Generation (SHG). SHG detects arrays of molecules arranged in a semi-crystalline order. Myosin filaments are a good source of SHG and the technique can be applied to large bundles of fibers. We have found that SHG is capable of detecting all the inclusions (enlarged lysosomes and autophagic buildup areas) found in the mouse model and in human biopsies of Pompe disease subjects (Ralston et al., 2008). This was the first published application of SHG to assess muscle disease. SHG quantitation of muscle defects should be also of interest in other pathological conditions in which structural defects of muscles may be more subtle and not as well documented.